Abstract District Heating Networks (DHNs) are crucial to decarbonizing the heat supply sector, with the evolution toward 4th and 5th generation systems offering significant potential for efficiency. However, the increasing complexity of these modern systems makes high-fidelity dynamic thermo-hydraulic simulation computationally intensive, particularly for large-scale networks. These simulations are essential for key applications such as thermal loss prediction, supply temperature optimization, operational planning and during the sizing phase of the network. Recent research has utilized Machine Learning (ML)-based surrogate models to replace substation clusters, reducing spatial complexity and accelerating simulations. Yet, the efficacy of this spatial reduction is highly sensitive to the clusters definition. Poorly selected clusters degrade the surrogate models accuracy and undermine the simulation performance. This paper proposes a flexible, task-driven graph clustering methodology specifically designed for ML-based spatial reduction. We introduce physics-informed distance metrics that encode the primary drivers of ML surrogate model errors. These distance metrics are leveraged within a hierarchical agglomerative clustering framework. The methodology is evaluated across 16 DHNs with diverse topological and thermal characteristics. Our results reveal a clear, generalizable trade-off between simulation accuracy and computational time reduction, allowing the integration of user-preferences. On an independent validation DHN, a high physical accuracy preference yielded a deviation of 6.45 MWh in total thermal energy production estimation (3.76% of total thermal losses within the network), albeit with modest computational gains. Expressing the error relative to thermal losses, rather than total production, provides a more meaningful measure of the spatial reduction’s impact on the key operational objectives. Conversely, a high spatial reduction preference reduced the network from 123 to 5 remaining physical nodes and achieved a 91% decrease in computational time, at the cost of a 45.24 MWh deviation in total thermal energy production estimation, representing 26.37% error relative to thermal losses. These findings demonstrate that the proposed clustering framework provides a robust and flexible tool to calibrate the balance between physical fidelity of the simulation and computational speed across varying DHN generations.

Multiple criteria decision aiding helps decision-makers (DMs) to reach better decisions in multi-criteria problems using preference models, whose parameters are elicited to best correspond to the preferences of the DM using, among other things, holistic judgments. Regarding the elicitation of the Reference based on Multiple reference Profiles (RMP) model, the literature only contains an exact method based on a Boolean satisfiability formulation, while a mixed-integer linear program and evolutionary metaheuristics focus on a more simpler version of the model (SRMP). Exact methods for preference elicitation usually struggle to solve cases with many criteria and a lot of preference information, while metaheuristics have a gap to optimal solutions. To address these two issues, we propose in this article a simulated annealing based metaheuristic to elicit an RMP model. In order to evaluate the performance of this method, we conducted numerical experiments on simulated instances. To this end, we developed a way to uniformly generate a weak-order extension on the subsets of criteria, which allows us to create random DMs consistent with the RMP model. The results of these experiments show that the proposed method is able to solve at optimality big instances, as well as being closer to optimal solutions than other metaheuristics in SRMP elicitation.

District Heating Networks (DHNs) offer a sustainable approach to thermal energy distribution by integrating low-carbon heat sources. However, their inherent complexity calls for optimized strategies that balance operational costs, user comfort, and capital investment. Rapid scenario assessment-enabled by reliable digital twins and efficient simulation tools-is essential to support such optimization. A key limitation in many existing simulation models is the lack of validation against real-world measurements. This paper presents a comparative validation of two fundamentally different quasi-dynamic simulation models, HeatGrid and PyDHN, using on-site measurements from a real-world meshed DHN. Despite their distinct methodologies, both models produced closely aligned results, with mean absolute differences of only 0.0268 kg/s in mass flow rate and 0.75. C in return temperature at the heating station. In addition to validation, the models were then used to evaluate the economic potential of adding new looping pipes scenarios to the network. Simulations revealed that although the additional pipes reduced pumping energy, they also led to significantly higher heat transport losses. As a result, the cost savings from reduced pumping energy were outweighed by the increased thermal losses-rendering all pipe addition scenarios economically unviable. Overall, the use of the prevalidated simulation models enabled rapid and informed evaluation of extension scenarios. The findings highlight that, contrary to intuition, adding pipes does not necessarily lead to lower operating costs in meshed DHNs.

District heating networks (DHNs) provide an efficient heat distribution solution in urban areas, accomplished through interconnected and insulated pipes linking local heat sources to local consumers. This efficiency is further enhanced by the capacity of these networks to integrate renewable heat sources and thermal storage systems. However, integration of these systems adds complexity to the physical dynamics of the network, necessitating complex dynamic simulation models. These dynamic physical simulations are computationally expensive, limiting their adoption, particularly in large-scale networks. To address this challenge, we propose a methodology utilizing Artificial Neural Networks (ANNs) to reduce the computational time associated with the DHNs dynamic simulations. Our approach consists in replacing predefined clusters of substations within the DHNs with trained surrogate ANNs models, effectively transforming these clusters into single nodes. This creates a hybrid simulation framework combining the predictions of the ANNs models with the accurate physical simulations of remaining substation nodes and pipes. We evaluate different architectures of Artificial Neural Network on diverse clusters from four synthetic DHNs with realistic heating demands. Results demonstrate that ANNs effectively learn cluster dynamics irrespective of topology or heating demand levels. Through our experiments, we achieved a 27% reduction in simulation time by replacing 39% of consumer nodes while maintaining acceptable accuracy in preserving the generated heat powers by sources.

This work proposes a Large Neighborhood Search Metaheuristic for solving a mixed-model assembly line balancing problem with walking workers and dynamic task assignment. The considered problem is a multi-stage stochastic program with integer recourse. These problems are very hard to solve because the number of binary variables increases exponentially with the number of production cycles. We study different decomposition approaches, and our results suggest that re-optimizing for a sub-tree outperforms other decompositions, such as model-based or station decomposition.

Evaluating the quality of data is a problem of a multi-dimensional nature and quite frequently depends on the perspective of an expected use or final purpose of the data. Numerous works have explored the well-known specification of data quality dimensions in various application domains, without addressing the inter-dependencies and aggregation of quality attributes for decision support. In this work we therefore propose a context-dependent formal process to evaluate the quality of data which integrates a preference model from the field of Multi-Criteria Decision Aiding. The parameters of this preference model are determined through interviews with work-domain experts. We show the interest of the proposal on a case study related to the evaluation of the quality of hydrographical survey data.

Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Copy DOI

In the context of Multiple Criteria Decision Aiding, decision makers often face problems with multiple conflicting criteria that justify the use of preference models to help advancing towards a decision. In order to determine the parameters of these preference models, preference elicitation makes use of preference learning algorithms, usually taking as input holistic judgments, i.e., overall preferences on some of the alternatives, expressed by the decision maker. Tools to achieve this goal in the context of a ranking model based on multiple reference profiles are usually based on mixed-integer linear programming, Boolean satisfiability formulation or metaheuristics. However, they are usually unable to handle realistic problems involving many criteria and a large amount of input information. We propose here an evolutionary metaheuristic in order to address this issue. Extensive experiments illustrate its ability to handle problem instances that previous proposals cannot.

Multiple Criteria Decision Aiding (MCDA) provides preference models and algorithms to assist decision-makers (DMs) in their decision-making tasks. The preference models are characterized by preference parameters which can be learned through preference learning algorithms from holistic judgments given by the DM. Here, we use Simulated Annealing (SA) to learn the parameters of the Ranking based on Multiple Reference Profiles (RMP) model and its simpler variant SRMP. Extensive experiments demonstrate that our proposal outperforms existing methods in terms of both calculation time and accuracy.

SRMP, which stands for “Simple Ranking with Multiple Profiles”, is a Multi-Criteria Decision Aiding model aiming to rank alternatives according to the preferences of a Decision Maker (DM) using reference criteria evaluations. Determining the preference parameters of SRMP can be tiring for the DM, who is often asked to compare several alternatives pairwisely during a preference elicitation process. It has been proposed in the literature to use an incremental elicitation process which selects informative pairs of alternatives which are submitted to the DM in sequence. The goal in such a process is to refine the SRMP model at each iteration, until a robust recommendation is determined, while limiting the cognitive effort of the DM. In this research, using a regret-based elicitation approach, we present a new heuristic for choosing the pairs of alternatives sequentially submitted for evaluation to the DM. We also provide a mixed-integer linear program for an efficient computation of regret values in practice. We limit our solution to the elicitation of the criteria weights, a subset of the SRMP model’s parameters, and we demonstrate that in this setting, the suggested heuristic outperforms previously examined query selection algorithms.

The emergence of the SARS-CoV-2 virus and new viral variations with higher transmission and mortal- ity rates have highlighted the urgency to accelerate vaccination to mitigate the morbidity and mortality of the COVID-19 pandemic. For this purpose, this paper formulates a new multi-vaccine, multi-depot location-inventory-routing problem for vaccine distribution. The proposed model addresses a wide variety of vaccination concerns: prioritizing age groups, fair distribution, multi-dose injection, dynamic demand, etc. To solve large-size instances of the model, we employ a Benders decomposition algorithm with a number of acceleration techniques. To monitor the dynamic demand of vaccines, we propose a new ad- justed susceptible-infectious-recovered (SIR) epidemiological model, where infected individuals are tested and quarantined. The solution to the optimal control problem dynamically allocates the vaccine demand to reach the endemic equilibrium point. Finally, to illustrate the applicability and performance of the proposed model and solution approach, the paper reports extensive numerical experiments on a real case study of the vaccination campaign in France. The computational results show that the proposed Benders decomposition algorithm is 12 times faster, and its solutions are, on average, 16% better in terms of qual- ity than the Gurobi solver under a limited CPU time. In terms of vaccination strategies, our results suggest that delaying the recommended time interval between doses of injection by a factor of 1.5 reduces the unmet demand up to 50%. Furthermore, we observed that the mortality is a convex function of fairness and an appropriate level of fairness should be adapted through the vaccination.

This paper aims at integrating machine learning techniques into meta-heuristics for solving combinato-rial optimization problems. Specifically, our study develops a novel efficient iterated greedy algorithm based on reinforcement learning. The main novelty of the proposed algorithm is its new perturbation mechanism, which incorporates Q-learning to select appropriate perturbation operators during the search process. Through an application to the permutation flowshop scheduling problem, comprehensive com-putational experiments are conducted on a wide range of benchmark instances to evaluate the perfor-mance of the proposed algorithm. This evaluation is done against non-learning versions of the iterated greedy algorithm and seven state-of-the-art algorithms from the literature. The experimental results and statistical analyses show the better performance of the proposed algorithm in terms of optimality gaps, convergence rate, and computational overhead.(c) 2022 Elsevier B.V. All rights reserved.

Among decision problems in spatial management planning, marine spatial planning (MSP) has lately gained popularity. One of the difficulties in MSP is to determine the best place for a new activity while taking into account the locations of current activities. This paper presents the results of the extension of one multi-objective evolutionary-based algorithm (MOEA), non-dominated sorting genetic algorithm-II (NSGA-II) solved the multi-objective spatial zoning optimization problem. The proposed algorithm aims to maximize the interest of the area of the zone dedicated to the new activity while maximizing its spatial compactness. The extended NSGA-II, unlike the traditional one, makes use of a different stop condition, four crossover operators, three mutation operators, and repairing operators. This algorithm is developed for the raster data and it computes solutions for the multi-objective spatial zoning optimization model at a large scale. The proposed NSGA-II has revealed a good performance in comparison with the exact method tested on a small scale. To improve the performance of the algorithm, its parameters are calibrated and tuned using the Multi-Response Surface Methodology (MRSM) method. Analysis of variance (ANOVA) was used to determine the effective and non-effective factors and correctness of the regression models. Finally, conclusions are made and future research works are recommended.

The COVID-19 virus's high transmissibility has resulted in the virus's rapid spread throughout the world, which has brought several repercussions, ranging from a lack of sanitary and medical products to the collapse of medical systems. Hence, governments attempt to re-plan the production of medical products and reallocate limited health resources to combat the pandemic. This paper addresses a multi-period production-inventory-sharing problem (PISP) to overcome such a circumstance, considering two consumable and reusable products. We introduce a new formulation to decide on production, inventory, delivery, and sharing quantities. The sharing will depend on net supply balance, allowable demand overload, unmet demand, and the reuse cycle of reusable products. Undeniably, the dynamic demand for products during pandemic situations must be reflected effectively in addressing the multi-period PISP. A bespoke compartmental susceptible-exposed-infectious-hospitalized-recovered-susceptible (SEIHRS) epidemiological model with a control policy is proposed, which also accounts for the influence of people's behavioral response as a result of the knowledge of adequate precautions. An accelerated Benders decomposition-based algorithm with tailored valid inequalities is offered to solve the model. Finally, we consider a realistic case study - the COVID-19 pandemic in France - to examine the computational proficiency of the decomposition method. The computational results reveal that the proposed decomposition method coupled with effective valid inequalities can solve large-sized test problems in a reasonable computational time and 9.88 times faster than the commercial Gurobi solver. Moreover, the sharing mechanism reduces the total cost of the system and the unmet demand on the average up to 32.98% and 20.96%, respectively.

The optimization of multi-energy systems (MESs) in which multiple energy carriers interact with each other is a complex problem.Their optimal operation and design can be determined through mathematical programming.A classical technology used in MESs is the combined heat and power units (CHP) whose efficiency is modeled through non-linear equations.These non-linear functions are approximated through piecewise linear ones by introducing binary decision variables, which generates a mixed-integer linear program (MILP).Consequently, optimizing such systems over a long time period with a high temporal resolution becomes infeasible in a reasonable amount of time.In this work, we propose a fast heuristic algorithm to optimize the design and operation of such an MES.Our case study is an MES at the scale of a district with five types of generation units, including a CHP, over a time period of one year with a temporal resolution of one hour.Comparison of the proposed heuristic and a state-of-the-art MILP solver over smaller time periods shows that the heuristic is up to 99.9 % faster, with a mean error of 2.3 × 10 -4 % compared to the optimal solution.The heuristic can also solve the optimal design and operation problem over a year in about 10 minutes.

The emergence of the SARS-CoV-2 virus and new viral variations with higher transmission and mortality rates have highlighted the urgency to accelerate vaccination to mitigate the morbidity and mortality of the COVID-19 pandemic. For this purpose, this paper formulates a new multi-vaccine, multi-depot location-inventory-routing problem for vaccine distribution. The proposed model addresses a wide variety of vaccination concerns: prioritizing age groups, fair distribution, multi-dose injection, dynamic demand, etc. To solve large-size instances of the model, we employ a Benders decomposition algorithm with a number of acceleration techniques. To monitor the dynamic demand of vaccines, we propose a new adjusted susceptible-infectious-recovered (SIR) epidemiological model, where infected individuals are tested and quarantined. The solution to the optimal control problem dynamically allocates the vaccine demand to reach the endemic equilibrium point. Finally, to illustrate the applicability and performance of the proposed model and solution approach, the paper reports extensive numerical experiments on a real case study of the vaccination campaign in France. The computational results show that the proposed Benders decomposition algorithm is 12 times faster, and its solutions are, on average, 16% better in terms of quality than the Gurobi solver under a limited CPU time. In terms of vaccination strategies, our results suggest that delaying the recommended time interval between doses of injection by a factor of 1.5 reduces the unmet demand up to 50%. Furthermore, we observed that the mortality is a convex function of fairness and an appropriate level of fairness should be adapted through the vaccination.

Sewerage systems need to be improved in order to respect new regulatory frameworks which should lower the impacts on the environment and allow to prepare for increases of water discharge. In this paper, we propose a method to evaluate sewerage systems improvement scenarios through multiple criteria by using a Multi-Criteria Decision Aiding process. In doing so, we facilitate the integration of the different points of view of the various stakeholders and decision maker(s) in this evaluation. Our proposal also eases the process of the study, by proposing a clear, interactive and iterative business process, which allows all project actors to be integrated in an explicit way. Since the proposed solution is transparent, the final recommendation is more easily adopted by the different actors. We illustrate the genericity of the proposed approach through a real case study in the city of Brest in France and show how this work helped the project participants adopt the proposed recommendations.

Multi-energy systems (MESs) combining different energy carriers like electricity and heat allow for more efficient and sustainable energy solutions. However, optimizing the design and operation of MESs is challenging due to non-linearities in the mathematical models used, especially the performance curves of technologies like combined heat and power units. Unlike similar work from the literature, this paper proposes an improved piecewise linearization method to efficiently handle the non-linearities, models an MES as a multi-objective mixed-integer linear program (MILP), and solves the optimization problem over a year with hourly resolution to enable detailed operation and faithful system design. The method uses fewer linear pieces to approximate non-linear functions compared to a standard technique, resulting in lower complexity while preserving accuracy. The MES design and operation problem maximizes cost reduction and the rate of renewable energy sources. A case study on an MES with electricity and heat over one year with hourly resolution demonstrates the effectiveness of the new method. It allows for solving a long-term MES optimization problem in reasonable computation times.